Reply

Abstract

To the Editor:We are glad for the opportunity to respond to issues raised by Newman and Maisels.1Johnson LH Brown AK Bhutani VK. System-based approach to management of neonatal jaundice and prevention of kernicterus.J Pediatr. 2002; 140: 396-403Abstract Full Text Full Text PDF PubMed Scopus (256) Google Scholar These include questions about the significance and reliability of reports of kernicterus in healthy infants born since the mid-1980s, compared with its near eliminations during the preceding 20 years. Our concern about the occurrence of kernicterus in healthy term newborns was raised in 1991.2Johnson LH. Hyperbilirubinemia in the term and near-term infant. When to worry, when to treat.N Y State J Med. 1991; 91: 483-489PubMed Google Scholar This was one year before publication of the paper by Newman and Maisels, “Management of jaundice in the term newborn: A kinder, gentler appprach,”3Newman TB Maisels MJ. Does hyperbilirubinemia damage the brain of healthy full-term infants?.Clin Perinatol. 1990; 17: 331-358PubMed Google Scholar and one year after their article, “Does hyperbilirubinemia damage the brain of healthy term infants?”4Newman TB Maisels MJ. Evaluation and treatment of jaundice in the term newborn: a kinder, gentler approach.Pediatrics. 1992; 89: 809-818PubMed Google Scholar Invited commentaries for the “…kinder, gentler…” article by other bilirubin “experts” were published in the same 1992 issue of Pediatrics.5Valaes T Wennber R Poland R Cashore W Gartner L Brown AK et al.Invited commentaries to “…kinder, gentler approach…”4.Pediatrics. 1992; 89: 819-833PubMed Google Scholar In several of these commentaries, concern was expressed that adoption of new recommendations without further scrutiny might put infants at increased risk for adverse outcomes secondary to hyperbilirubinemia and that no mechanism had been proposed to evaluate their safety and efficacy. In a written response to these commentaries (Pediatrics 1992;89:831), Newman and Maisels agreed “…that the evidence on which we based our recommendations is not sufficient to generate a new “standard of care” for jaundiced infants. We believe, however, that our recommendations are more consistent with the available (imperfect) data that the previous recommendations were…We join Drs Cashore and Wennberg in encouraging groups like the American Academy of Pediatrics (AAP) to develop their own practice guidelines, and that, whatever guidelines are developed, outcome evaluation will be an important step in the process. In the meantime, we believe the kinder, gentler approach to the jaundiced infant is also the more prudent.” In a recent article,6Newman TB Maisels MJ. Less aggressive treatment of neonatal jaundice and reports of kernicterus: lessons about practice guidelines.Pediatrics. 2000; 105: 242-245Crossref PubMed Scopus (28) Google Scholar Newman and Maisels have reiterated this opinion. No formal evaluation of whether their “kinder, gentler” approach is indeed more prudent, in an era of shortened hospital stays, cost containment and advocacy of breast-feeding, has been initiated either by the authors or by the AAP.Analysis of data presented in our paper argues strongly that kernicterus has re-emerged in the United States and that management of jaundice needs to be more rigorous, more broadly based, and easier to implement and monitor. Our system-based approach relies on the best available evidence and, in our studies, has been found to be efficient, cost-effective, and most importantly, safer for all newborns.1Johnson LH Brown AK Bhutani VK. System-based approach to management of neonatal jaundice and prevention of kernicterus.J Pediatr. 2002; 140: 396-403Abstract Full Text Full Text PDF PubMed Scopus (256) Google Scholar, 7Bhutani VK, Johnson LH. System-based approach to the management of newborn jaundice to prevent kernicterus. CD-ROM (Ross Laboratories). Presented at Hot Topics Symposium, 2001, Washington, DC.Google Scholar, 8Khurana E Bhutani VK Dworanczyk R Spitz D Sivieri EM Johnson LH. System-based management of neonatal jaundice and impact on readmission for severe hyperbilirubinemia.Pediatr Res. 2002; 51: 322aGoogle Scholar It deserves futher scrutiny.Unknown risk of unmonitored jaundice. It is necessary to point out the importance as well as the limitations of Newman et al's study of the incidence of “extreme” hyperbilirubinemia in the “captive” newborn population of a mature health maintenance organization.9Newman TB Xiong B Gonzales VM Escobar GJ. Prediction and prevention of extreme neonatal hyperbilirubinemia in a mature health maintenance organization.Arch Pediatr Adolesc Med. 2000; 154: 1140-1147Crossref PubMed Scopus (164) Google Scholar Follow-up of these newborns after hospital discharge was an established policy, as was general adherence to AAP guidelines for phototherapy. Newman et al reported an incidence of 1:700 for bilirubin levels of ≥25 mg/dL in healthy newborns (birth weight >2000 g, gestational age ≥36 weeks) in spite of such preventive measures and the exclusion of infants with early-onset jaundice who were treated before hospital discharge. Not addressed in such studies is the incidence of dangerous hyperbilirubinemia in the population at large who receive diverse, sometimes inadequate, healthcare services.Continuing occurrence of neonatal mortality and morbidity due to kernicterus. Ongoing cases of kernicterus continue to be reported to the Registry. Only those enrolled as of January 2001 were included in our paper. Among cases reported after January 2001 are 3 additional instances of fatal kernicterus. These 3 infants were born in metropolitan centers of 3 different states, one in 2000, one in 2001, and one in 2002.Response to specific queries. The numerator reported in our paper is the number of cases found eligible for the Registry, among infants born (n-90) from 1984 to January 2001 from 27 of the United States. As of August 2002, the Registry includes cases (n-110) from 31 states plus a US Army camp in Germany. All of these cases were preventable.1Johnson LH Brown AK Bhutani VK. System-based approach to management of neonatal jaundice and prevention of kernicterus.J Pediatr. 2002; 140: 396-403Abstract Full Text Full Text PDF PubMed Scopus (256) Google Scholar The denominator for reported cases is unknown. Newman and Maisels imply that the denominator should be all healthy newborns. Kernicterus occurs in infants discharged as well as from the newborn nursery, whether they are term or near-term, with or without risk factors. It might be more meaningful to use a denominator that approximates infants with potentially high-risk hyperbilirubinemia (>75th percentile for age in hours).Numerator accuracy. The number of reports to the Pilot Registry is an underestimate, because kernicterus is an “underground” disease in the United States. The existence of the Registry and requests for confidential reports have been made yearly since 1992 at the Annual Kernicterus Symposium (PAS meetings). There have also been ongoing, interim reports of data from the Registry in abstract form and in a Year Book of Neonatal and Perinatal Medicine article in 1996.1Johnson LH Brown AK Bhutani VK. System-based approach to management of neonatal jaundice and prevention of kernicterus.J Pediatr. 2002; 140: 396-403Abstract Full Text Full Text PDF PubMed Scopus (256) Google Scholar All cases were contributed by colleagues and/or requested consultations (both professional and medico-legal). All met the eligibility criteria for enrollment and were individually evaluated to prevent duplication. Eligibility requirements, rules for identification of patients, for patient and professional confidentiality, and requests for input and contributions from the medical community are stated in the manuscript. Only cases of “definite” kernicterus were enrolled. No infant or child who had nonspecific or minor impairment was included.Incidence of kernicterus. No US data are or will be available and applicable to current practic unless an ethically unfeasible prospective study is conducted. However, the comprehensive vital statistics available in Denmark suggest that there indeed has been a reemergence of kernicterus in our era.10Ebbesen F. Recurrence of kernicterus in term and near-term infants in Denmark.Acta Pediatrica. 2000; 89: 1213-1217Crossref PubMed Google Scholar Societal expectation in the United States is that kernicterus should not occur in any healthy newborn. Must we wait to introduce and evaluate safer practices for a disease that should have a zero occurrence?Definition. Classic clinical signs of acute kernicterus are diagnostic, as are the recognized clinical signs of “prethreshold” and “threshold” retinopathy of prematurity. Both diseases need prompt intervention to prevent or minimize long-term sequelae. Our paper describes six babies with unequivocal acute stage kernicterus, who received treatment with intensive phototherapy, exchange transfusion, and sometimes serum albumin infusions and appeared to have escaped irreversible bilirubin brain damage. Two of these infants had severe progressive signs of acute stage damage. They were not expected to have good outcomes. Their peak total serum bilirubin (TSB) levels were 30.6 and 25.6 mg/dL, respectively. The other 4 infants had classic acute stage signs but of lesser severity. The peak bilirubin levels in 3 of the 4 were 29.5 mg/dL, 30 mg/dL, and 35 mg/dL. The peak TSB in the fourth was not reported but a neonatologist, experienced in the diagnosis of acute and chronic kernicterus, documented the presence of unequivocal acute bilirubin encephalopathy.With regard to the proposition that a specific TSB level be required for diagnosis of kernicterus, the median TSB level in the cases of kernicterus reported in our paper was >35 mg/dL. Nevertheless, it is essential to remember that infants with TSB levels <30 mg/dL before the age of 72 hours and <25 mg/dL before the age of 36 to 48 hours are almost certainly at as great a risk as infants with higher levels at a later age. These estimates are based on changes in albumin bilirubin binding affinity with postnatal age, changes in blood brain barrier function, vascular integrity, and blood gas equilibria in the first 24 to 48 hours after birth, as well as clinical experience. there were a total of 7 babies with peak TSB levels <30 mg/dL at readmission by age 7 days who had classic signs of acute kernicterus. One infant was lost to follow-up, 4 had severe kernicteric sequelae, and 2 seem to have escaped irreversible damage.As we balance evidence-based medicine with patient safety, we, as pediatricians, need to be truly prudent and protective of all newborns entrusted to our care.YMPD71 To the Editor:We are glad for the opportunity to respond to issues raised by Newman and Maisels.1Johnson LH Brown AK Bhutani VK. System-based approach to management of neonatal jaundice and prevention of kernicterus.J Pediatr. 2002; 140: 396-403Abstract Full Text Full Text PDF PubMed Scopus (256) Google Scholar These include questions about the significance and reliability of reports of kernicterus in healthy infants born since the mid-1980s, compared with its near eliminations during the preceding 20 years. Our concern about the occurrence of kernicterus in healthy term newborns was raised in 1991.2Johnson LH. Hyperbilirubinemia in the term and near-term infant. When to worry, when to treat.N Y State J Med. 1991; 91: 483-489PubMed Google Scholar This was one year before publication of the paper by Newman and Maisels, “Management of jaundice in the term newborn: A kinder, gentler appprach,”3Newman TB Maisels MJ. Does hyperbilirubinemia damage the brain of healthy full-term infants?.Clin Perinatol. 1990; 17: 331-358PubMed Google Scholar and one year after their article, “Does hyperbilirubinemia damage the brain of healthy term infants?”4Newman TB Maisels MJ. Evaluation and treatment of jaundice in the term newborn: a kinder, gentler approach.Pediatrics. 1992; 89: 809-818PubMed Google Scholar Invited commentaries for the “…kinder, gentler…” article by other bilirubin “experts” were published in the same 1992 issue of Pediatrics.5Valaes T Wennber R Poland R Cashore W Gartner L Brown AK et al.Invited commentaries to “…kinder, gentler approach…”4.Pediatrics. 1992; 89: 819-833PubMed Google Scholar In several of these commentaries, concern was expressed that adoption of new recommendations without further scrutiny might put infants at increased risk for adverse outcomes secondary to hyperbilirubinemia and that no mechanism had been proposed to evaluate their safety and efficacy. In a written response to these commentaries (Pediatrics 1992;89:831), Newman and Maisels agreed “…that the evidence on which we based our recommendations is not sufficient to generate a new “standard of care” for jaundiced infants. We believe, however, that our recommendations are more consistent with the available (imperfect) data that the previous recommendations were…We join Drs Cashore and Wennberg in encouraging groups like the American Academy of Pediatrics (AAP) to develop their own practice guidelines, and that, whatever guidelines are developed, outcome evaluation will be an important step in the process. In the meantime, we believe the kinder, gentler approach to the jaundiced infant is also the more prudent.” In a recent article,6Newman TB Maisels MJ. Less aggressive treatment of neonatal jaundice and reports of kernicterus: lessons about practice guidelines.Pediatrics. 2000; 105: 242-245Crossref PubMed Scopus (28) Google Scholar Newman and Maisels have reiterated this opinion. No formal evaluation of whether their “kinder, gentler” approach is indeed more prudent, in an era of shortened hospital stays, cost containment and advocacy of breast-feeding, has been initiated either by the authors or by the AAP.Analysis of data presented in our paper argues strongly that kernicterus has re-emerged in the United States and that management of jaundice needs to be more rigorous, more broadly based, and easier to implement and monitor. Our system-based approach relies on the best available evidence and, in our studies, has been found to be efficient, cost-effective, and most importantly, safer for all newborns.1Johnson LH Brown AK Bhutani VK. System-based approach to management of neonatal jaundice and prevention of kernicterus.J Pediatr. 2002; 140: 396-403Abstract Full Text Full Text PDF PubMed Scopus (256) Google Scholar, 7Bhutani VK, Johnson LH. System-based approach to the management of newborn jaundice to prevent kernicterus. CD-ROM (Ross Laboratories). Presented at Hot Topics Symposium, 2001, Washington, DC.Google Scholar, 8Khurana E Bhutani VK Dworanczyk R Spitz D Sivieri EM Johnson LH. System-based management of neonatal jaundice and impact on readmission for severe hyperbilirubinemia.Pediatr Res. 2002; 51: 322aGoogle Scholar It deserves futher scrutiny.Unknown risk of unmonitored jaundice. It is necessary to point out the importance as well as the limitations of Newman et al's study of the incidence of “extreme” hyperbilirubinemia in the “captive” newborn population of a mature health maintenance organization.9Newman TB Xiong B Gonzales VM Escobar GJ. Prediction and prevention of extreme neonatal hyperbilirubinemia in a mature health maintenance organization.Arch Pediatr Adolesc Med. 2000; 154: 1140-1147Crossref PubMed Scopus (164) Google Scholar Follow-up of these newborns after hospital discharge was an established policy, as was general adherence to AAP guidelines for phototherapy. Newman et al reported an incidence of 1:700 for bilirubin levels of ≥25 mg/dL in healthy newborns (birth weight >2000 g, gestational age ≥36 weeks) in spite of such preventive measures and the exclusion of infants with early-onset jaundice who were treated before hospital discharge. Not addressed in such studies is the incidence of dangerous hyperbilirubinemia in the population at large who receive diverse, sometimes inadequate, healthcare services.Continuing occurrence of neonatal mortality and morbidity due to kernicterus. Ongoing cases of kernicterus continue to be reported to the Registry. Only those enrolled as of January 2001 were included in our paper. Among cases reported after January 2001 are 3 additional instances of fatal kernicterus. These 3 infants were born in metropolitan centers of 3 different states, one in 2000, one in 2001, and one in 2002.Response to specific queries. The numerator reported in our paper is the number of cases found eligible for the Registry, among infants born (n-90) from 1984 to January 2001 from 27 of the United States. As of August 2002, the Registry includes cases (n-110) from 31 states plus a US Army camp in Germany. All of these cases were preventable.1Johnson LH Brown AK Bhutani VK. System-based approach to management of neonatal jaundice and prevention of kernicterus.J Pediatr. 2002; 140: 396-403Abstract Full Text Full Text PDF PubMed Scopus (256) Google Scholar The denominator for reported cases is unknown. Newman and Maisels imply that the denominator should be all healthy newborns. Kernicterus occurs in infants discharged as well as from the newborn nursery, whether they are term or near-term, with or without risk factors. It might be more meaningful to use a denominator that approximates infants with potentially high-risk hyperbilirubinemia (>75th percentile for age in hours).Numerator accuracy. The number of reports to the Pilot Registry is an underestimate, because kernicterus is an “underground” disease in the United States. The existence of the Registry and requests for confidential reports have been made yearly since 1992 at the Annual Kernicterus Symposium (PAS meetings). There have also been ongoing, interim reports of data from the Registry in abstract form and in a Year Book of Neonatal and Perinatal Medicine article in 1996.1Johnson LH Brown AK Bhutani VK. System-based approach to management of neonatal jaundice and prevention of kernicterus.J Pediatr. 2002; 140: 396-403Abstract Full Text Full Text PDF PubMed Scopus (256) Google Scholar All cases were contributed by colleagues and/or requested consultations (both professional and medico-legal). All met the eligibility criteria for enrollment and were individually evaluated to prevent duplication. Eligibility requirements, rules for identification of patients, for patient and professional confidentiality, and requests for input and contributions from the medical community are stated in the manuscript. Only cases of “definite” kernicterus were enrolled. No infant or child who had nonspecific or minor impairment was included.Incidence of kernicterus. No US data are or will be available and applicable to current practic unless an ethically unfeasible prospective study is conducted. However, the comprehensive vital statistics available in Denmark suggest that there indeed has been a reemergence of kernicterus in our era.10Ebbesen F. Recurrence of kernicterus in term and near-term infants in Denmark.Acta Pediatrica. 2000; 89: 1213-1217Crossref PubMed Google Scholar Societal expectation in the United States is that kernicterus should not occur in any healthy newborn. Must we wait to introduce and evaluate safer practices for a disease that should have a zero occurrence?Definition. Classic clinical signs of acute kernicterus are diagnostic, as are the recognized clinical signs of “prethreshold” and “threshold” retinopathy of prematurity. Both diseases need prompt intervention to prevent or minimize long-term sequelae. Our paper describes six babies with unequivocal acute stage kernicterus, who received treatment with intensive phototherapy, exchange transfusion, and sometimes serum albumin infusions and appeared to have escaped irreversible bilirubin brain damage. Two of these infants had severe progressive signs of acute stage damage. They were not expected to have good outcomes. Their peak total serum bilirubin (TSB) levels were 30.6 and 25.6 mg/dL, respectively. The other 4 infants had classic acute stage signs but of lesser severity. The peak bilirubin levels in 3 of the 4 were 29.5 mg/dL, 30 mg/dL, and 35 mg/dL. The peak TSB in the fourth was not reported but a neonatologist, experienced in the diagnosis of acute and chronic kernicterus, documented the presence of unequivocal acute bilirubin encephalopathy.With regard to the proposition that a specific TSB level be required for diagnosis of kernicterus, the median TSB level in the cases of kernicterus reported in our paper was >35 mg/dL. Nevertheless, it is essential to remember that infants with TSB levels <30 mg/dL before the age of 72 hours and <25 mg/dL before the age of 36 to 48 hours are almost certainly at as great a risk as infants with higher levels at a later age. These estimates are based on changes in albumin bilirubin binding affinity with postnatal age, changes in blood brain barrier function, vascular integrity, and blood gas equilibria in the first 24 to 48 hours after birth, as well as clinical experience. there were a total of 7 babies with peak TSB levels <30 mg/dL at readmission by age 7 days who had classic signs of acute kernicterus. One infant was lost to follow-up, 4 had severe kernicteric sequelae, and 2 seem to have escaped irreversible damage.As we balance evidence-based medicine with patient safety, we, as pediatricians, need to be truly prudent and protective of all newborns entrusted to our care.YMPD71 We are glad for the opportunity to respond to issues raised by Newman and Maisels.1Johnson LH Brown AK Bhutani VK. System-based approach to management of neonatal jaundice and prevention of kernicterus.J Pediatr. 2002; 140: 396-403Abstract Full Text Full Text PDF PubMed Scopus (256) Google Scholar These include questions about the significance and reliability of reports of kernicterus in healthy infants born since the mid-1980s, compared with its near eliminations during the preceding 20 years. Our concern about the occurrence of kernicterus in healthy term newborns was raised in 1991.2Johnson LH. Hyperbilirubinemia in the term and near-term infant. When to worry, when to treat.N Y State J Med. 1991; 91: 483-489PubMed Google Scholar This was one year before publication of the paper by Newman and Maisels, “Management of jaundice in the term newborn: A kinder, gentler appprach,”3Newman TB Maisels MJ. Does hyperbilirubinemia damage the brain of healthy full-term infants?.Clin Perinatol. 1990; 17: 331-358PubMed Google Scholar and one year after their article, “Does hyperbilirubinemia damage the brain of healthy term infants?”4Newman TB Maisels MJ. Evaluation and treatment of jaundice in the term newborn: a kinder, gentler approach.Pediatrics. 1992; 89: 809-818PubMed Google Scholar Invited commentaries for the “…kinder, gentler…” article by other bilirubin “experts” were published in the same 1992 issue of Pediatrics.5Valaes T Wennber R Poland R Cashore W Gartner L Brown AK et al.Invited commentaries to “…kinder, gentler approach…”4.Pediatrics. 1992; 89: 819-833PubMed Google Scholar In several of these commentaries, concern was expressed that adoption of new recommendations without further scrutiny might put infants at increased risk for adverse outcomes secondary to hyperbilirubinemia and that no mechanism had been proposed to evaluate their safety and efficacy. In a written response to these commentaries (Pediatrics 1992;89:831), Newman and Maisels agreed “…that the evidence on which we based our recommendations is not sufficient to generate a new “standard of care” for jaundiced infants. We believe, however, that our recommendations are more consistent with the available (imperfect) data that the previous recommendations were…We join Drs Cashore and Wennberg in encouraging groups like the American Academy of Pediatrics (AAP) to develop their own practice guidelines, and that, whatever guidelines are developed, outcome evaluation will be an important step in the process. In the meantime, we believe the kinder, gentler approach to the jaundiced infant is also the more prudent.” In a recent article,6Newman TB Maisels MJ. Less aggressive treatment of neonatal jaundice and reports of kernicterus: lessons about practice guidelines.Pediatrics. 2000; 105: 242-245Crossref PubMed Scopus (28) Google Scholar Newman and Maisels have reiterated this opinion. No formal evaluation of whether their “kinder, gentler” approach is indeed more prudent, in an era of shortened hospital stays, cost containment and advocacy of breast-feeding, has been initiated either by the authors or by the AAP. Analysis of data presented in our paper argues strongly that kernicterus has re-emerged in the United States and that management of jaundice needs to be more rigorous, more broadly based, and easier to implement and monitor. Our system-based approach relies on the best available evidence and, in our studies, has been found to be efficient, cost-effective, and most importantly, safer for all newborns.1Johnson LH Brown AK Bhutani VK. System-based approach to management of neonatal jaundice and prevention of kernicterus.J Pediatr. 2002; 140: 396-403Abstract Full Text Full Text PDF PubMed Scopus (256) Google Scholar, 7Bhutani VK, Johnson LH. System-based approach to the management of newborn jaundice to prevent kernicterus. CD-ROM (Ross Laboratories). Presented at Hot Topics Symposium, 2001, Washington, DC.Google Scholar, 8Khurana E Bhutani VK Dworanczyk R Spitz D Sivieri EM Johnson LH. System-based management of neonatal jaundice and impact on readmission for severe hyperbilirubinemia.Pediatr Res. 2002; 51: 322aGoogle Scholar It deserves futher scrutiny. Unknown risk of unmonitored jaundice. It is necessary to point out the importance as well as the limitations of Newman et al's study of the incidence of “extreme” hyperbilirubinemia in the “captive” newborn population of a mature health maintenance organization.9Newman TB Xiong B Gonzales VM Escobar GJ. Prediction and prevention of extreme neonatal hyperbilirubinemia in a mature health maintenance organization.Arch Pediatr Adolesc Med. 2000; 154: 1140-1147Crossref PubMed Scopus (164) Google Scholar Follow-up of these newborns after hospital discharge was an established policy, as was general adherence to AAP guidelines for phototherapy. Newman et al reported an incidence of 1:700 for bilirubin levels of ≥25 mg/dL in healthy newborns (birth weight >2000 g, gestational age ≥36 weeks) in spite of such preventive measures and the exclusion of infants with early-onset jaundice who were treated before hospital discharge. Not addressed in such studies is the incidence of dangerous hyperbilirubinemia in the population at large who receive diverse, sometimes inadequate, healthcare services. Continuing occurrence of neonatal mortality and morbidity due to kernicterus. Ongoing cases of kernicterus continue to be reported to the Registry. Only those enrolled as of January 2001 were included in our paper. Among cases reported after January 2001 are 3 additional instances of fatal kernicterus. These 3 infants were born in metropolitan centers of 3 different states, one in 2000, one in 2001, and one in 2002. Response to specific queries. The numerator reported in our paper is the number of cases found eligible for the Registry, among infants born (n-90) from 1984 to January 2001 from 27 of the United States. As of August 2002, the Registry includes cases (n-110) from 31 states plus a US Army camp in Germany. All of these cases were preventable.1Johnson LH Brown AK Bhutani VK. System-based approach to management of neonatal jaundice and prevention of kernicterus.J Pediatr. 2002; 140: 396-403Abstract Full Text Full Text PDF PubMed Scopus (256) Google Scholar The denominator for reported cases is unknown. Newman and Maisels imply that the denominator should be all healthy newborns. Kernicterus occurs in infants discharged as well as from the newborn nursery, whether they are term or near-term, with or without risk factors. It might be more meaningful to use a denominator that approximates infants with potentially high-risk hyperbilirubinemia (>75th percentile for age in hours). Numerator accuracy. The number of reports to the Pilot Registry is an underestimate, because kernicterus is an “underground” disease in the United States. The existence of the Registry and requests for confidential reports have been made yearly since 1992 at the Annual Kernicterus Symposium (PAS meetings). There have also been ongoing, interim reports of data from the Registry in abstract form and in a Year Book of Neonatal and Perinatal Medicine article in 1996.1Johnson LH Brown AK Bhutani VK. System-based approach to management of neonatal jaundice and prevention of kernicterus.J Pediatr. 2002; 140: 396-403Abstract Full Text Full Text PDF PubMed Scopus (256) Google Scholar All cases were contributed by colleagues and/or requested consultations (both professional and medico-legal). All met the eligibility criteria for enrollment and were individually evaluated to prevent duplication. Eligibility requirements, rules for identification of patients, for patient and professional confidentiality, and requests for input and contributions from the medical community are stated in the manuscript. Only cases of “definite” kernicterus were enrolled. No infant or child who had nonspecific or minor impairment was included. Incidence of kernicterus. No US data are or will be available and applicable to current practic unless an ethically unfeasible prospective study is conducted. However, the comprehensive vital statistics available in Denmark suggest that there indeed has been a reemergence of kernicterus in our era.10Ebbesen F. Recurrence of kernicterus in term and near-term infants in Denmark.Acta Pediatrica. 2000; 89: 1213-1217Crossref PubMed Google Scholar Societal expectation in the United States is that kernicterus should not occur in any healthy newborn. Must we wait to introduce and evaluate safer practices for a disease that should have a zero occurrence? Definition. Classic clinical signs of acute kernicterus are diagnostic, as are the recognized clinical signs of “prethreshold” and “threshold” retinopathy of prematurity. Both diseases need prompt intervention to prevent or minimize long-term sequelae. Our paper describes six babies with unequivocal acute stage kernicterus, who received treatment with intensive phototherapy, exchange transfusion, and sometimes serum albumin infusions and appeared to have escaped irreversible bilirubin brain damage. Two of these infants had severe progressive signs of acute stage damage. They were not expected to have good outcomes. Their peak total serum bilirubin (TSB) levels were 30.6 and 25.6 mg/dL, respectively. The other 4 infants had classic acute stage signs but of lesser severity. The peak bilirubin levels in 3 of the 4 were 29.5 mg/dL, 30 mg/dL, and 35 mg/dL. The peak TSB in the fourth was not reported but a neonatologist, experienced in the diagnosis of acute and chronic kernicterus, documented the presence of unequivocal acute bilirubin encephalopathy. With regard to the proposition that a specific TSB level be required for diagnosis of kernicterus, the median TSB level in the cases of kernicterus reported in our paper was >35 mg/dL. Nevertheless, it is essential to remember that infants with TSB levels <30 mg/dL before the age of 72 hours and <25 mg/dL before the age of 36 to 48 hours are almost certainly at as great a risk as infants with higher levels at a later age. These estimates are based on changes in albumin bilirubin binding affinity with postnatal age, changes in blood brain barrier function, vascular integrity, and blood gas equilibria in the first 24 to 48 hours after birth, as well as clinical experience. there were a total of 7 babies with peak TSB levels <30 mg/dL at readmission by age 7 days who had classic signs of acute kernicterus. One infant was lost to follow-up, 4 had severe kernicteric sequelae, and 2 seem to have escaped irreversible damage. As we balance evidence-based medicine with patient safety, we, as pediatricians, need to be truly prudent and protective of all newborns entrusted to our care. YMPD71